A Recurrent and Meta-learned Model of Weakly Supervised Object Localization


Sariyildiz M. B., Sumbul G., CİNBİŞ R. G.

6th International Symposium on Multidisciplinary Studies and Innovative Technologies, ISMSIT 2022, Ankara, Turkey, 20 - 22 October 2022, pp.747-752 identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/ismsit56059.2022.9932735
  • City: Ankara
  • Country: Turkey
  • Page Numbers: pp.747-752
  • Keywords: meta learning, recurrent neural networks, Weakly supervised object localization
  • Middle East Technical University Affiliated: Yes

Abstract

© 2022 IEEE.The object localization and detection has improved greatly over the past decade, thanks to developments in deep learning based representations and localization models. However, a major bottleneck remains at the reliance on fully-supervised datasets, which can be difficult to gather in many real-world scenarios. In this work, we focus on the problem of weakly-supervised localization, where the goal is to localize instances of objects based on simple image-level class annotations. In particular, instead of engineering a specific weakly-supervised localization model, we aim to meta-learn a recurrent neural network based model that aims to take series of training images of a novel class, and progressively discover the foreground pattern over them. We experimentally explore the model over scenes composed of MNIST digits and noisy patches as distractors.